tegrakernel/kernel/kernel-4.9/drivers/net/wireless/intel/iwlwifi/mvm/offloading.c

256 lines
8.4 KiB
C

/******************************************************************************
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
* Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
* Copyright(c) 2015 Intel Deutschland GmbH
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110,
* USA
*
* The full GNU General Public License is included in this distribution
* in the file called COPYING.
*
* Contact Information:
* Intel Linux Wireless <linuxwifi@intel.com>
* Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
*
* BSD LICENSE
*
* Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
* Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
* Copyright(c) 2015 Intel Deutschland GmbH
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* * Neither the name Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*****************************************************************************/
#include <net/ipv6.h>
#include <net/addrconf.h>
#include <linux/bitops.h>
#include "mvm.h"
void iwl_mvm_set_wowlan_qos_seq(struct iwl_mvm_sta *mvm_ap_sta,
struct iwl_wowlan_config_cmd *cmd)
{
int i;
/*
* For QoS counters, we store the one to use next, so subtract 0x10
* since the uCode will add 0x10 *before* using the value while we
* increment after using the value (i.e. store the next value to use).
*/
for (i = 0; i < IWL_MAX_TID_COUNT; i++) {
u16 seq = mvm_ap_sta->tid_data[i].seq_number;
seq -= 0x10;
cmd->qos_seq[i] = cpu_to_le16(seq);
}
}
int iwl_mvm_send_proto_offload(struct iwl_mvm *mvm,
struct ieee80211_vif *vif,
bool disable_offloading,
bool offload_ns,
u32 cmd_flags)
{
union {
struct iwl_proto_offload_cmd_v1 v1;
struct iwl_proto_offload_cmd_v2 v2;
struct iwl_proto_offload_cmd_v3_small v3s;
struct iwl_proto_offload_cmd_v3_large v3l;
} cmd = {};
struct iwl_host_cmd hcmd = {
.id = PROT_OFFLOAD_CONFIG_CMD,
.flags = cmd_flags,
.data[0] = &cmd,
.dataflags[0] = IWL_HCMD_DFL_DUP,
};
struct iwl_proto_offload_cmd_common *common;
u32 enabled = 0, size;
u32 capa_flags = mvm->fw->ucode_capa.flags;
#if IS_ENABLED(CONFIG_IPV6)
struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
int i;
/*
* Skip tentative address when ns offload is enabled to avoid
* violating RFC4862.
* Keep tentative address when ns offload is disabled so the NS packets
* will not be filtered out and will wake up the host.
*/
bool skip_tentative = offload_ns;
if (capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL ||
capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_LARGE) {
struct iwl_ns_config *nsc;
struct iwl_targ_addr *addrs;
int n_nsc, n_addrs;
int c;
int num_skipped = 0;
if (capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL) {
nsc = cmd.v3s.ns_config;
n_nsc = IWL_PROTO_OFFLOAD_NUM_NS_CONFIG_V3S;
addrs = cmd.v3s.targ_addrs;
n_addrs = IWL_PROTO_OFFLOAD_NUM_IPV6_ADDRS_V3S;
} else {
nsc = cmd.v3l.ns_config;
n_nsc = IWL_PROTO_OFFLOAD_NUM_NS_CONFIG_V3L;
addrs = cmd.v3l.targ_addrs;
n_addrs = IWL_PROTO_OFFLOAD_NUM_IPV6_ADDRS_V3L;
}
/*
* For each address we have (and that will fit) fill a target
* address struct and combine for NS offload structs with the
* solicited node addresses.
*/
for (i = 0, c = 0;
i < mvmvif->num_target_ipv6_addrs &&
i < n_addrs && c < n_nsc; i++) {
struct in6_addr solicited_addr;
int j;
if (skip_tentative &&
test_bit(i, mvmvif->tentative_addrs)) {
num_skipped++;
continue;
}
addrconf_addr_solict_mult(&mvmvif->target_ipv6_addrs[i],
&solicited_addr);
for (j = 0; j < c; j++)
if (ipv6_addr_cmp(&nsc[j].dest_ipv6_addr,
&solicited_addr) == 0)
break;
if (j == c)
c++;
addrs[i].addr = mvmvif->target_ipv6_addrs[i];
addrs[i].config_num = cpu_to_le32(j);
nsc[j].dest_ipv6_addr = solicited_addr;
memcpy(nsc[j].target_mac_addr, vif->addr, ETH_ALEN);
}
if (mvmvif->num_target_ipv6_addrs - num_skipped)
enabled |= IWL_D3_PROTO_IPV6_VALID;
if (capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL)
cmd.v3s.num_valid_ipv6_addrs =
cpu_to_le32(i - num_skipped);
else
cmd.v3l.num_valid_ipv6_addrs =
cpu_to_le32(i - num_skipped);
} else if (capa_flags & IWL_UCODE_TLV_FLAGS_D3_6_IPV6_ADDRS) {
bool found = false;
BUILD_BUG_ON(sizeof(cmd.v2.target_ipv6_addr[0]) !=
sizeof(mvmvif->target_ipv6_addrs[0]));
for (i = 0; i < min(mvmvif->num_target_ipv6_addrs,
IWL_PROTO_OFFLOAD_NUM_IPV6_ADDRS_V2); i++) {
if (skip_tentative &&
test_bit(i, mvmvif->tentative_addrs))
continue;
memcpy(cmd.v2.target_ipv6_addr[i],
&mvmvif->target_ipv6_addrs[i],
sizeof(cmd.v2.target_ipv6_addr[i]));
found = true;
}
if (found) {
enabled |= IWL_D3_PROTO_IPV6_VALID;
memcpy(cmd.v2.ndp_mac_addr, vif->addr, ETH_ALEN);
}
} else {
bool found = false;
BUILD_BUG_ON(sizeof(cmd.v1.target_ipv6_addr[0]) !=
sizeof(mvmvif->target_ipv6_addrs[0]));
for (i = 0; i < min(mvmvif->num_target_ipv6_addrs,
IWL_PROTO_OFFLOAD_NUM_IPV6_ADDRS_V1); i++) {
if (skip_tentative &&
test_bit(i, mvmvif->tentative_addrs))
continue;
memcpy(cmd.v1.target_ipv6_addr[i],
&mvmvif->target_ipv6_addrs[i],
sizeof(cmd.v1.target_ipv6_addr[i]));
found = true;
}
if (found) {
enabled |= IWL_D3_PROTO_IPV6_VALID;
memcpy(cmd.v1.ndp_mac_addr, vif->addr, ETH_ALEN);
}
}
if (offload_ns && (enabled & IWL_D3_PROTO_IPV6_VALID))
enabled |= IWL_D3_PROTO_OFFLOAD_NS;
#endif
if (capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL) {
common = &cmd.v3s.common;
size = sizeof(cmd.v3s);
} else if (capa_flags & IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_LARGE) {
common = &cmd.v3l.common;
size = sizeof(cmd.v3l);
} else if (capa_flags & IWL_UCODE_TLV_FLAGS_D3_6_IPV6_ADDRS) {
common = &cmd.v2.common;
size = sizeof(cmd.v2);
} else {
common = &cmd.v1.common;
size = sizeof(cmd.v1);
}
if (vif->bss_conf.arp_addr_cnt) {
enabled |= IWL_D3_PROTO_OFFLOAD_ARP | IWL_D3_PROTO_IPV4_VALID;
common->host_ipv4_addr = vif->bss_conf.arp_addr_list[0];
memcpy(common->arp_mac_addr, vif->addr, ETH_ALEN);
}
if (!disable_offloading)
common->enabled = cpu_to_le32(enabled);
hcmd.len[0] = size;
return iwl_mvm_send_cmd(mvm, &hcmd);
}